SecProbe.io

CORRESP
1
filename1.htm

                                                                                                                                                                                                                                         Southern First
Bancshares, Inc.

                                                                                                                                                                                                      100 Verdae Boulevard

                                                                                                                                                                                                                                                                    Greenville,
SC 29607

August 6, 2010

VIA EDGAR
CORRESPONDENCE

Mr. Amit Pande

Accounting
Branch Chief

Division
of Corporation Finance

Securities
and Exchange Commission

100
F Street, NE

Washington,
DC 20549

          Re:      Southern
First Bancshares, Inc.

                      File
No. 000-27719

                      Form
10-K for the fiscal year ended December 31, 2009

                      Form
10-Q for the quarterly period ended March 31, 2010

Dear Mr.
Pande:

This
letter responds to your letter dated July 23, 2010 to Southern First
Bancshares, Inc. (the “Company”). For your convenience, we have reprinted the
Staff’s numbered comments below followed by the Company’s responses. Per our
conversation with Mr. Ben Phippen, we have included responses as of the most
recent period available, which is June 30, 2010.  In addition, corresponding
language reflecting the responses below will be included in the Company’s Form
10-Q for the period ended June 30, 2010, which will be filed with the SEC no
later than next week.

Form 10-Q for
the quarterly period ended March 31, 2010

Management's
Discussion and Analysis of Financial Condition and Results of Operations

Investments,
page 22

1.      Please provide us with, and revise future filings to include,
a table similar to that provided on page 62 of your December 31, 2009 Form 10-K
detailing the amortized cost, gross unrealized gains, gross unrealized losses
and fair value for each category of investments for which other than temporary
impairments have not been recognized in earnings (including investments for
which a portion of an other-than-temporary has been recognized in other
comprehensive income). Also provide us with, and revise future filings to
include, a table similar to that provided on page 63 of your December 31, 2009
Form l0-K detailing the information requested in the preceding sentence
segregated by those investments that have been in a continuous uurea1ized loss
position for less than 12 months and those that have been in a continuous
uurea1ized loss position for 12 months or longer. Refer to ASC 320-1 0-50-6a
and ASC 320-10-50-7 in preparing your response and note that this disclosure is
required in all annual and quarterly filings.

Response:
The amortized costs and fair value of investment securities
available for sale and held to maturity are as follows (dollars in thousands):

	June 30, 2010

	Amortized

	Gross Unrealized

	Fair

	Cost

	Gains

	Losses

	Value

	Available for
  sale

	Government
  sponsored enterprises

	 $

	4,500

	 $

	-

	 $

	2

	 $

	4,498

	State and
  political subdivisions

	7,992

	93

	10

	8,075

	12,492

	93

	12

	12,573

	Mortgage-backed
  securities

	FHLMC

	24,660

	289

	45

	24,904

	FNMA

	39,730

	644

	-

	40,374

	GNMA

	1,183

	90

	-

	1,273

	Collateralized
  mortgage obligations

	6,427

	-

	350

	6,077

	72,000

	1,023

	395

	72,628

	Total

	$

	84,492

	$

	1,116

	$

	407

	$

	85,201

	Held to
  maturity

	Mortgage-backed
  securities

	FNMA

	 $

	120

	 $

	5

	 $

	-

	 $

	125

The table below
summarizes gross unrealized losses on investment securities and the fair market
value of the related securities, aggregated by investment category and length
of time that individual securities have been in a continuous unrealized loss
position, at June 30, 2010 (dollars in thousands).

	Less than 12 months

	12 months or longer

	Total

	Fair

	Unrealized

	Fair

	Unrealized

	Fair

	Unrealized

	#

	value

	losses

	#

	value

	losses

	#

	value

	losses

	As of June 30,
  2010

	Available for
  sale

	Government
  sponsored enterprises

	1

	$

	4,498

	$

	2

	-

	$

	-

	$

	-

	1

	$

	4,498

	$

	2

	State and
  political subdivision

	3

	2,111

	10

	-

	-

	-

	3

	2,111

	10

	Mortgage-backed

	FHLMC

	2

	4,477

	45

	-

	-

	-

	2

	4,477

	45

	Collateral
  mortgage obligations

	-

	-

	-

	1

	2,890

	350

	1

	2,890

	350

	6

	$

	11,086

	$

	57

	1

	$

	2,890

	$

	350

	7

	$

	13,976

	$

	407

2.       As a related matter, please refer to ASC 320-10-50-6b to
provide us with, and revise future filings to include, an enhanced discussion
of your private label collateralized mortgage obligations that are in a
continuous loss position. Your response and the revised disclosure in future
filings should provide sufficient information to allow financial statement
users to understand the quantitative disclosures and the information that you
considered (both positive and negative) in reaching the conclusion that the
impairments are not other-than-temporary. For example, please consider
disclosing additional information to describe the characteristics of the
underlying collateral (e.g., types of mortgages, vintage, geographic
concentrations, average FICO scores, etc), performance indicators of the
underlying assets (e.g., quantification of default rates, relevant delinquency
and foreclosure statistics and trends, etc), quantification of excess
collateral, changes in credit ratings, and quantification of any changes in the
major components of your internal cash flow model that occurred during the
periods presented.

2

Response:
The primary risk in conforming mortgage backed securities is
prepayment risk.  Credit risk is assumed by the government sponsored
enterprises.  In collateralized non-agency mortgage obligations, credit risk is
prominent.  Credit risk, arising from borrower default, is especially difficult
to quantify given the market’s lack of clarity on future home prices as well as
the impact of loan attributes.  Credit risk is of paramount concern to
investors that own collateralized mortgage obligations (“CMOs”).

At June 30,
2010, we held two CMOs with a book value of $6.4 million, one of which was in
an unrealized loss position at June 30, 2010. The Company evaluates these
securities quarterly based on the methodology outlined below and based on this
evaluation, believes that it will receive all of the principal and interest in
accordance with the original contractual terms of the securities. Also, since the
company has no intent to sell securities with unrealized losses and it is not
more-likely-than-not that we will be required to sell these securities before
recovery of amortized cost, we have concluded that the securities are not
impaired on an other-than-temporary basis.

Evaluation
Methodology

The primary
cause of mortgage delinquency and foreclosure is attributable to the loss of
household income due to unemployment.  Therefore, the change in unemployment
rates is a predictor of the likelihood of mortgage loan delinquency and
foreclosures.  Prime mortgage borrowers, with secure or steady employment tend
to stay current on their mortgages, even if home prices drop significantly, as
in a period of severe home price depreciation.  Subprime borrowers, on the
other hand, who only marginally sustained their initial home purchase, are more
likely to become delinquent when home prices dropped precipitously.  There is
evidence to support the premise that increases in unemployment rates and home
price depreciation have a positive correlation with foreclosure rates.

In the non
agency MBS sector, FICO scores that measure the credit worthiness of the
mortgage borrower are a good indicator of future mortgage delinquency and
foreclosure rates. In addition, the level of documentation that was obtained at
the time the loan was originated is also a strong indicator of future loan
performance. Loans that have limited or reduced documentation have proven to
result in higher delinquency and foreclosures. An additional indicator of the
likelihood of delinquency and foreclosures is the occupancy type.  Generally
speaking, loans on owner occupied properties tend to be less likely to default
than loans on vacation or investment properties. Listed below is various
historical data related to our two private label CMOs.

 CMO-
A                                CMO-B

Book value on
6-30-2010 (in thousands)                       $3,187                                   $3,240

Year of
origination                                                              2006                                       2006

Original credit
rating                                                            Aaa,AAA                              Aaa,AAA

Current credit
rating                                                            Ba1,BBB-                             Caa1,CC

Original credit
support                                                        15.62                                      13.48

Current credit
support                                                         13.88
7.48

Current coverage
ratio                                                          1.8
.6

Percentage of
loans 60+ delinquent                                 22.8%                                    38.3%

Severity loss
rate - 12 months                                           53%                                        48%

Constant
prepayment rate:

      3 months                                                                         14.7%                                    21.5%

      6 months                                                                         13.4                                        15.5

    12 months                                                                         14.1                                        14.8

Weighted
average:

     Coupon                                                                            6.24%                                    6.14%

     Months
remaining                                                          309                                         307

     Loan size                                                                          $301,000                               $510,000

Current loan to value                                                      79%                                        72%

FICO scores:

   Original                                                                               736                                         716

   Current                                                                               734                                         713

Current
percentage of limited documents                       38%                                        63%

Current
percentage – owner occupied                             89%                                        81%

Geographic
concentration:

                Highest
percentage                                             GA 77.9%                             CA
45.6%

                Second
highest percentage                                FL  15.8%                             NY
17.3%

3

Our private
label non-agency securities CMOs were priced based on Bloomberg Default model
due to market illiquidity.  We noted that this market has had little, if any,
new issuance since the credit crisis began.  The company determined that most
sales are forced and do not reflect the true economic value of these
securities.  The Bloomberg default model is comprised of five sub-models:

		•       Prepayment
models

		•       Foreclosure
models

		•       Liquidity
models

		•       Loss
severity models

		•       60+
Delinquency rate models

The prepayment
and foreclosure models consider loan level inputs including loan to value, loan
size, FICO score, loan purpose, loan type, property type, loan documentation,
state of origin, and weighted average coupon.  In addition, the model includes
macro-economic inputs such as interest rate, mortgage rates, home price appreciation
and unemployment rates based on the various metropolitan areas.  Regression
analysis including logistical and survival analysis are used for model
development.

The liquidation
model projections are derived from net foreclosure rates and the transition
matrices.  The liquidation rates are calculated separately according to
judicial and non-judicial states.

The loss
severity models are calculated based on each loan characteristics and other
macro-economic inputs such as:

		•       Original
loan to value

		•       Cumulative
home price appreciation

		•       Liquidation
markdown

		•       Back
interest owed

		•       Property
tax owed

		•       Broker
and lawyer fees

Although, the
individual loan level data was utilized as of June 30, 2010, the home price
appreciation assumptions and the future unemployment rate assumptions were as
of December 1, 2009.  The current home price appreciation assumptions are as
follows:

		•       Continue
to decline at an annual rate of 10% till December of 2009

		•       Stays
at 0% for two years

		•       Increases
at an annual rate of 5% for two years

		•       Increases
at an annual rate of 3% for the remaining periods

The unemployment
rate assumptions are as follows:

		•       Continues
to rise to 10.5% till January of 2010

		•       Stay
at 10.5% for 3 months

		•       Declines
to 5.8% over the next 5 years

		•       Continue
at 5.8% for the remaining periods

Based on the
independent calculations and assumptions, management currently anticipates
receiving all of the outstanding principal and the related interest for each
security. Management has  reviewed the independent assumptions utilized, compared
them to current actual results and  believes that they are reasonable.  However,
there is no precise method to predict if credit losses in the future periods
will exceed our current predictions.  If actual results significantly vary from
the assumptions noted above, we may be required to recognize losses that are
later deemed to not be only temporary in nature.  The valuation change has been
recorded as a change in the unrealized gain/loss recognized in other
comprehensive income.

4

3.      We note your disclosure that you have the ability and intent
to hold all securities within the portfolio until they mature or until the
value recovers and therefore you do not consider these investments to be
other-than-temporarily impaired at March 31, 2010. In future filings, please
revise this language to be consistent with ASC 320-10-35-33, which clarified
the interaction of factors that should be considered when determining whether a
debt security is other-than-temporarily impaired from an assertion of intent
and ability to hold a security for a period of time sufficient to allow for an
anticipated recovery in fair value to an assessment of whether (a) one has the
intent to sell the security, or (b) it is more likely than not that one will be
required to sell the security prior to its anticipated recovery.

Response:
On June 30, 2010, the Company had 6 individual investments
that were in an unrealized loss position.  The unrealized losses were primarily
attributable to changes in interest rates, rather than deterioration in credit
quality.  The majority of these securities are government or agency securities
and, therefore, pose minimal credit risk.  The company considers the length of
time and extent to which the fair value of AFS debt securities have been less
than cost to conclude that such securities were not other-than-temporarily
impaired. We also consider other factors such as the